Texture-Aware Fast Global Level Set Evolution

This work presents a two phase image segmentation strategy using the level set method.

Due to its intrinsic advantages such as the ability to automatically handle complex shapes and topological changes, the level set method has been widely used in image segmentation. Nevertheless, in addition to be computational expensive, it has the limitation to very often lead to a local minimum because of the energy functional to be minimized is non-convex. In this work, we use the geometric active contours and the image thresholding frameworks to design a novel method for global image segmentation. The local lattice Boltzmann method is used to solve the level set equation. The proposed algorithm is therefore effective and highly parallelizable. Experimental results on satellite, natural and medical images demonstrate the effectiveness and the efficiency of the proposed method when implemented using an NVIDIA graphics processing units

The present Matlab code is the implementation of our work.
For using:

1- Open the main code (TAFGLSE.m),
2- Put the adress of the image you want to segment,
3- Run the code (optional : add to path or change the folder).

If you use this code, please cite the implemented paper as follow

S. Balla-Arabé, X.-B Gao and L. Xu, “Texture-Aware Fast Global Level Set Evolution,” 4th International Conference on Intelligence Science and Big Data Engineering, Springer LNCS 8261, Beijing, 2013.

If you have any question please let me know.

E -mail: [e-mail]
Google Scholar link: http://scholar.google.fr/citations?user=dYEIx_IAAAAJ&hl=en